Google Local Search Update: New Ranking Factors for 2026

What Actually Changed in Local Search This Autumn

If you run a local business and your phone has felt quieter — or busier — than your rankings would suggest, you are not imagining things. Over the back half of 2025, the local results have shifted in ways that do not announce themselves with a Google blog post. There was no single dated update labelled the way core updates are. Instead, practitioners watching dozens of markets at once have been comparing notes on a cluster of changes that became hard to ignore around the start of October: review signals behaving differently, the map pack taking up less room on some queries and more on others, service-area businesses being treated with new scepticism, and a visible acceleration in spam enforcement against keyword-stuffed business names.

A word of caution before any of it. Nobody outside Google knows the ranking factors, and nobody knows their weights. Anyone selling you a leaked list of percentages is selling fiction. What follows is not a leak. It is a read of observed patterns — what local SEO professionals are seeing repeatedly enough, across enough different businesses, that it is worth adjusting for. Treat each shift below as a direction the system appears to be moving in, not a setting you can dial. The value is not in the gossip; it is in the concrete adjustment each pattern points to, most of which you can make this week.

The through-line for 2026 is bigger than the map. The same structured local signals that decide pack position are now the raw material that AI assistants draw on when someone asks them, conversationally, to recommend a business nearby. So tightening these signals is no longer just about three slots on a map. It is about being the answer a machine is willing to give.

Reviews Are Being Read, Not Just Counted

The oldest piece of local-SEO folk wisdom is to get more reviews. It is still true, but it is no longer the whole picture, and the gap between count and quality has widened noticeably this year. Businesses with large but stale review profiles — a few hundred reviews, none newer than last spring — have been watching newer competitors with smaller, fresher profiles pull level or ahead. The pattern points to recency and velocity carrying more practical weight than a high lifetime total on its own.

This is the difference between a review profile that proves you were good once and one that proves you are good now. A steady trickle of recent reviews reads as a living business; a wall of old five-stars reads as a business coasting on past reputation, and the results increasingly seem to treat it that way. The corresponding pattern on velocity is just as clear: a sudden burst of reviews in a single week, after months of silence, looks less like success and more like manipulation, and the system has grown better at discounting bursts that do not match a business's normal rhythm.

There is also growing evidence that the words inside reviews are being parsed, not merely tallied. When customers mention the specific service and the specific place — the heat-pump install in the east end, the same-day denture repair — that language appears to help connect the listing to those searches in a way a bare star rating never could. You cannot and must not script this, but you can ask better. A request that nudges a customer to mention what you actually did for them produces a richer review than a request that asks only for stars.

The adjustment: stop treating reviews as a number to grow and start treating them as a cadence to maintain. Build a request into the close of every job, so fresh reviews arrive at a natural, steady rate. Audit your own profile for recency — if your newest review is months old, that is a ranking and trust problem hiding in plain sight, regardless of how high your total climbs.

Your Replies to Reviews Are a Surface, Not an Afterthought

Closely tied to the review shift is a quieter one around owner responses. Responding to reviews has always been good manners and good marketing. What practitioners are increasingly observing is that an actively managed response history correlates with stronger, steadier performance — and it makes intuitive sense that a profile where the owner replies to feedback looks more alive, and more trustworthy, than one where reviews vanish into silence.

The more interesting development is who the responses are really for. As AI systems summarize a business from its profile, your replies become part of the text they read. A calm, specific, helpful response to a critical review does double duty: it reassures the dozens of future customers who read it, and it gives an AI assistant evidence that the business handles problems like an adult. A defensive or absent response does the opposite on both counts.

The adjustment is undramatic but real. Respond to every review, positive and negative, and treat negative-review replies as public relations written for the next reader rather than rebuttals aimed at the reviewer. Acknowledge, state your side once without arguing, and move the specifics offline. The audience is everyone — human and machine — deciding whether you are a safe choice.

Proximity Is Still King, But the Throne Moved

Proximity — how close the searcher is to your business — remains the factor you cannot optimize away, and that has not changed. What has shifted is how proximity interacts with everything else, and the nuance matters because it is where most owners waste effort.

For years the simplification was that the nearest relevant business tends to win. The pattern emerging through 2025 is more textured. On highly competitive, high-intent queries, proximity still dominates: search from three blocks away and the closer business usually appears first. But on more specific or less crowded queries, a business slightly further away with a markedly stronger profile — better reviews, a more complete and relevant listing, more web prominence — has been overcoming a proximity disadvantage more often than the old rule of thumb would predict. Proximity is less of a hard gate and more of a heavy weight that genuine relevance and prominence can sometimes outpull.

The practical consequence most owners miss: this is exactly why checking your rank from your own desk tells you almost nothing. You are sitting on top of your own pin, so you always look like you are winning. Your customers are spread across a service area, seeing entirely different results. Rank tracking that samples many points across the area you actually serve is the only honest measurement, and it has become more important, not less, as proximity's behaviour grows more conditional.

The adjustment: win your immediate radius first, because near your address is where proximity is on your side and where the highest-intent searches happen. Then treat the edges of your area as a profile-strength and organic problem — somewhere a stronger listing can occasionally beat a closer competitor — rather than a battle you expect proximity to win for you. And replace desk-checking with multi-point tracking before you draw any conclusion about how you are doing.

Service-Area Businesses Are Under New Scrutiny

Businesses that travel to the customer rather than receiving them — plumbers, mobile detailers, cleaners, contractors — have always had a harder time in local search than storefront businesses, because the system is built around a physical pin. Through 2025 that gap has widened, and service-area businesses, or SABs, are reporting more volatility and more verification friction than their storefront counterparts.

Part of this is an anti-spam consequence. The SAB model, where the address is hidden and a service radius is declared instead, has been heavily abused by lead-generation operators creating phantom listings at addresses they do not occupy. The tightening that catches those operators also catches legitimate SABs in the net, which is why honest mobile businesses have seen more suspensions, reinstatement requests, and demands for proof of address this year. The defensible read is not that Google dislikes SABs — it is that the trust bar for a business with no public storefront has risen.

The pattern also rewards SABs that look unambiguously real. A profile with genuine photos of the team and the vehicles, a service area defined honestly rather than stretched across an entire region, consistent business details everywhere they appear, and a steady stream of recent reviews from the neighbourhoods actually served — that profile clears the higher trust bar that a thin, sprawling, generic SAB listing now trips over.

The adjustment: if you are an SAB, treat verifiability as a first-class task. Define a service area you can defend, not the biggest one you can claim. Keep your address documentation current in case you are asked to re-verify. Concentrate review-gathering and photos on proving you genuinely operate where you say you do. The businesses being hurt by the SAB tightening are overwhelmingly the ones that were stretching the truth; the cure is to be conspicuously legitimate.

The Crackdown on Keyword-Stuffed Names Finally Has Teeth

For years the most infuriating thing in local search was watching a competitor outrank you with a business name that was not their business name. A company legally called Riverside Dental listing itself as Riverside Dental - Best Emergency Dentist & Implants Toronto is breaking Google's guidelines, and because the name field carries genuine relevance weight, the cheat worked. Filing complaints felt like shouting into a void.

The news this year is that enforcement has become noticeably more responsive. Practitioners who report stuffed names through the proper channels are seeing more corrections, and faster ones, than in prior years — the suggest-an-edit function and the business redressal complaint form are no longer the dead letters they once seemed. None of this is instant or guaranteed, but the channel works often enough that policing your own market has gone from a long shot to a routine, worthwhile task.

There is a flip side that catches careless owners. The same tightening that removes competitors' stuffed names will remove yours. If at any point someone optimized your listing by appending keywords to the business name, that is now a liability rather than an edge, and a competitor can report it as easily as you can report theirs.

The adjustment is two-sided. Clean your own house first: your listed name should be your real-world name, exactly, with no bolted-on services or cities. Then audit the businesses outranking you. Where a stuffed name is doing the lifting, use suggest-an-edit to submit the correct one, and escalate persistent offenders through the redressal form. Every successful correction can hand you a slot you were losing to a rule you were following.

AI Answers Are Rewriting the Local Real Estate

The most consequential shift is the one that does not look like a local-search change at all. As AI-generated answers spread across the results in 2025, the space and the behaviour of local queries changed with them. On some searches the map pack sits lower, beneath an AI-composed summary; on others a conversational answer recommends a handful of businesses directly, before the map ever appears. The pixels available to win have moved, and they keep moving.

Where those AI answers get their information is the part that should reshape how you work. They are not crawling your homepage from scratch in the moment a question is asked. They lean on structured, verified local data, and the richest structured record of your business that exists is the one Google already holds — your profile, your reviews, your attributes — corroborated against what the wider web says about you. An assistant composing a recommendation uses your categories to decide whether you match the request, your review text to describe what you are good at, your attributes to answer constraint-based asks like open now or wheelchair accessible, and your overall prominence to decide whether you are a safe business to name.

This raises the stakes on everything above. A thin profile does not merely rank lower; it gives the assistant nothing to say about you, and assistants do not recommend businesses they cannot describe. Review text matters more, because a language model reads and summarizes it — a hundred reviews that repeatedly mention honest pricing and same-day service become, in effect, your machine-written reputation. Specific, well-described services matter more, because conversational queries are longer and more particular than typed ones. Consistency across the web matters more, because these systems cross-reference sources and quietly discount a business whose details disagree with themselves.

The adjustment is less a new task list than a reason to take the old one seriously. The encouraging part is that almost nobody in local categories is optimizing with this in mind yet. The business with the completest profile, the freshest reviews, the cleanest details, and the most specific service descriptions is positioned to win both the map pack of today and the recommendation engines steadily growing up around it.

What to Actually Do Before the Quarter Closes

None of these shifts demands a heroic project. They reward the unglamorous work most competitors skip, and they punish shortcuts that used to be safe. If you do five things in response to this autumn's changes, do these.

First, fix review recency. Build a review request into the close of every job so fresh reviews arrive steadily, and check that your newest review is recent — not impressive in total, recent. Second, respond to every review, writing negative replies for the next reader and the next machine, not the reviewer. Third, replace desk-checking with rank tracking that samples multiple points across your real service area, so you measure what your customers see rather than what your own location flatters you into believing. Fourth, if you are a service-area business, make yourself conspicuously legitimate: an honest service radius, current address documentation, real photos, and reviews from the places you actually serve. Fifth, audit business names — clean any stuffing from your own listing and report the stuffed names beating you, because that channel finally works.

Underneath all five is the same principle. Local search is becoming less forgiving of anything fake and more rewarding of anything verifiably real, because the verification is increasingly done by systems that also feed AI recommendations. The defensible strategy in 2026 is not to chase a rumoured ranking factor. It is to be, demonstrably and consistently, the legitimate local business the machine is looking for. If you want a second set of eyes on where your profile sits against your market, that is the kind of read SearchPod runs for local clients before recommending anything.